Random Forest Model
‘MyBank’ executed a campaign to cross-sell Personal Loans. As part of their Pilot Campaign, 20000 customers were sent campaigns through email , SMS , and direct mail. They all were given an offer of Personal Loan at an attractive interest rate of 12% and processing fee waived off if they respond within 1 Month. 2512 customer expressed their interest in that campaign.
Hence, ‘MyBank’ wants to find the profitable segments of the customers and target them to cross-sell their personal loans.
Loading the dataset
loan.data <- read.csv("D:/PGP BA-BI Course Materials/DATA MINING/GROUP ASSIGNMENT/PL_XSELL.csv",header = TRUE)Exploratory Data Analysis
Structure of the data
str(loan.data)## 'data.frame': 20000 obs. of 40 variables:
## $ CUST_ID : Factor w/ 20000 levels "C1","C10","C100",..: 17699 16532 11027 17984 2363 11747 18115 15556 15216 12494 ...
## $ TARGET : int 0 0 0 0 0 0 0 0 0 0 ...
## $ AGE : int 27 47 40 53 36 42 30 53 42 30 ...
## $ GENDER : Factor w/ 3 levels "F","M","O": 2 2 2 2 2 1 2 1 1 2 ...
## $ BALANCE : num 3384 287489 18217 71720 1671623 ...
## $ OCCUPATION : Factor w/ 4 levels "PROF","SAL","SELF-EMP",..: 3 2 3 2 1 1 1 2 3 1 ...
## $ AGE_BKT : Factor w/ 7 levels "<25",">50","26-30",..: 3 7 5 2 5 6 3 2 6 3 ...
## $ SCR : int 776 324 603 196 167 493 479 562 105 170 ...
## $ HOLDING_PERIOD : int 30 28 2 13 24 26 14 25 15 13 ...
## $ ACC_TYPE : Factor w/ 2 levels "CA","SA": 2 2 2 1 2 2 2 1 2 2 ...
## $ ACC_OP_DATE : Factor w/ 4869 levels "1/1/2000","1/1/2001",..: 2227 451 2689 3994 1515 3619 4286 2104 2009 2173 ...
## $ LEN_OF_RLTN_IN_MNTH : int 146 104 61 107 185 192 177 99 88 111 ...
## $ NO_OF_L_CR_TXNS : int 7 8 10 36 20 5 6 14 18 14 ...
## $ NO_OF_L_DR_TXNS : int 3 2 5 14 1 2 6 3 14 8 ...
## $ TOT_NO_OF_L_TXNS : int 10 10 15 50 21 7 12 17 32 22 ...
## $ NO_OF_BR_CSH_WDL_DR_TXNS: int 0 0 1 4 1 1 0 3 6 3 ...
## $ NO_OF_ATM_DR_TXNS : int 1 1 1 2 0 1 1 0 2 1 ...
## $ NO_OF_NET_DR_TXNS : int 2 1 1 3 0 0 1 0 4 0 ...
## $ NO_OF_MOB_DR_TXNS : int 0 0 0 1 0 0 0 0 1 0 ...
## $ NO_OF_CHQ_DR_TXNS : int 0 0 2 4 0 0 4 0 1 4 ...
## $ FLG_HAS_CC : int 0 0 0 0 0 1 0 0 1 0 ...
## $ AMT_ATM_DR : int 13100 6600 11200 26100 0 18500 6200 0 35400 18000 ...
## $ AMT_BR_CSH_WDL_DR : int 0 0 561120 673590 808480 379310 0 945160 198430 869880 ...
## $ AMT_CHQ_DR : int 0 0 49320 60780 0 0 10580 0 51490 32610 ...
## $ AMT_NET_DR : num 973557 799813 997570 741506 0 ...
## $ AMT_MOB_DR : int 0 0 0 71388 0 0 0 0 170332 0 ...
## $ AMT_L_DR : num 986657 806413 1619210 1573364 808480 ...
## $ FLG_HAS_ANY_CHGS : int 0 1 1 0 0 0 1 0 0 0 ...
## $ AMT_OTH_BK_ATM_USG_CHGS : int 0 0 0 0 0 0 0 0 0 0 ...
## $ AMT_MIN_BAL_NMC_CHGS : int 0 0 0 0 0 0 0 0 0 0 ...
## $ NO_OF_IW_CHQ_BNC_TXNS : int 0 0 0 0 0 0 0 0 0 0 ...
## $ NO_OF_OW_CHQ_BNC_TXNS : int 0 0 1 0 0 0 0 0 0 0 ...
## $ AVG_AMT_PER_ATM_TXN : num 13100 6600 11200 13050 0 ...
## $ AVG_AMT_PER_CSH_WDL_TXN : num 0 0 561120 168398 808480 ...
## $ AVG_AMT_PER_CHQ_TXN : num 0 0 24660 15195 0 ...
## $ AVG_AMT_PER_NET_TXN : num 486779 799813 997570 247169 0 ...
## $ AVG_AMT_PER_MOB_TXN : num 0 0 0 71388 0 ...
## $ FLG_HAS_NOMINEE : int 1 1 1 1 1 1 0 1 1 0 ...
## $ FLG_HAS_OLD_LOAN : int 1 0 1 0 0 1 1 1 1 0 ...
## $ random : num 1.14e-05 1.11e-04 1.20e-04 1.37e-04 1.74e-04 ...The given dataset has 20,000 customers data with 40 variables . We can also see the data type of each variable in the dataset.
Summary of the data
summary(loan.data)## CUST_ID TARGET AGE GENDER
## C1 : 1 Min. :0.0000 Min. :21.00 F: 5433
## C10 : 1 1st Qu.:0.0000 1st Qu.:30.00 M:14376
## C100 : 1 Median :0.0000 Median :38.00 O: 191
## C1000 : 1 Mean :0.1256 Mean :38.42
## C10000 : 1 3rd Qu.:0.0000 3rd Qu.:46.00
## C10001 : 1 Max. :1.0000 Max. :55.00
## (Other):19994
## BALANCE OCCUPATION AGE_BKT SCR
## Min. : 0 PROF :5417 <25 :1753 Min. :100.0
## 1st Qu.: 64754 SAL :5855 >50 :3035 1st Qu.:227.0
## Median : 231676 SELF-EMP:3568 26-30:3434 Median :364.0
## Mean : 511362 SENP :5160 31-35:3404 Mean :440.2
## 3rd Qu.: 653877 36-40:2814 3rd Qu.:644.0
## Max. :8360431 41-45:3067 Max. :999.0
## 46-50:2493
## HOLDING_PERIOD ACC_TYPE ACC_OP_DATE LEN_OF_RLTN_IN_MNTH
## Min. : 1.00 CA: 4241 11/16/2010: 24 Min. : 29.0
## 1st Qu.: 7.00 SA:15759 4/3/2009 : 23 1st Qu.: 79.0
## Median :15.00 7/25/2010 : 22 Median :125.0
## Mean :14.96 5/6/2013 : 21 Mean :125.2
## 3rd Qu.:22.00 2/7/2007 : 20 3rd Qu.:172.0
## Max. :31.00 8/24/2010 : 20 Max. :221.0
## (Other) :19870
## NO_OF_L_CR_TXNS NO_OF_L_DR_TXNS TOT_NO_OF_L_TXNS
## Min. : 0.00 Min. : 0.000 Min. : 0.00
## 1st Qu.: 6.00 1st Qu.: 2.000 1st Qu.: 9.00
## Median :10.00 Median : 5.000 Median : 14.00
## Mean :12.35 Mean : 6.634 Mean : 18.98
## 3rd Qu.:14.00 3rd Qu.: 7.000 3rd Qu.: 21.00
## Max. :75.00 Max. :74.000 Max. :149.00
##
## NO_OF_BR_CSH_WDL_DR_TXNS NO_OF_ATM_DR_TXNS NO_OF_NET_DR_TXNS
## Min. : 0.000 Min. : 0.000 Min. : 0.000
## 1st Qu.: 1.000 1st Qu.: 0.000 1st Qu.: 0.000
## Median : 1.000 Median : 1.000 Median : 0.000
## Mean : 1.883 Mean : 1.029 Mean : 1.172
## 3rd Qu.: 2.000 3rd Qu.: 1.000 3rd Qu.: 1.000
## Max. :15.000 Max. :25.000 Max. :22.000
##
## NO_OF_MOB_DR_TXNS NO_OF_CHQ_DR_TXNS FLG_HAS_CC AMT_ATM_DR
## Min. : 0.0000 Min. : 0.000 Min. :0.0000 Min. : 0
## 1st Qu.: 0.0000 1st Qu.: 0.000 1st Qu.:0.0000 1st Qu.: 0
## Median : 0.0000 Median : 2.000 Median :0.0000 Median : 6900
## Mean : 0.4118 Mean : 2.138 Mean :0.3054 Mean : 10990
## 3rd Qu.: 0.0000 3rd Qu.: 4.000 3rd Qu.:1.0000 3rd Qu.: 15800
## Max. :25.0000 Max. :15.000 Max. :1.0000 Max. :199300
##
## AMT_BR_CSH_WDL_DR AMT_CHQ_DR AMT_NET_DR AMT_MOB_DR
## Min. : 0 Min. : 0 Min. : 0 Min. : 0
## 1st Qu.: 2990 1st Qu.: 0 1st Qu.: 0 1st Qu.: 0
## Median :340150 Median : 23840 Median : 0 Median : 0
## Mean :378475 Mean : 124520 Mean :237308 Mean : 22425
## 3rd Qu.:674675 3rd Qu.: 72470 3rd Qu.:473971 3rd Qu.: 0
## Max. :999930 Max. :4928640 Max. :999854 Max. :199667
##
## AMT_L_DR FLG_HAS_ANY_CHGS AMT_OTH_BK_ATM_USG_CHGS
## Min. : 0 Min. :0.0000 Min. : 0.000
## 1st Qu.: 237936 1st Qu.:0.0000 1st Qu.: 0.000
## Median : 695115 Median :0.0000 Median : 0.000
## Mean : 773717 Mean :0.1106 Mean : 1.099
## 3rd Qu.:1078927 3rd Qu.:0.0000 3rd Qu.: 0.000
## Max. :6514921 Max. :1.0000 Max. :250.000
##
## AMT_MIN_BAL_NMC_CHGS NO_OF_IW_CHQ_BNC_TXNS NO_OF_OW_CHQ_BNC_TXNS
## Min. : 0.000 Min. :0.00000 Min. :0.0000
## 1st Qu.: 0.000 1st Qu.:0.00000 1st Qu.:0.0000
## Median : 0.000 Median :0.00000 Median :0.0000
## Mean : 1.292 Mean :0.04275 Mean :0.0444
## 3rd Qu.: 0.000 3rd Qu.:0.00000 3rd Qu.:0.0000
## Max. :170.000 Max. :2.00000 Max. :2.0000
##
## AVG_AMT_PER_ATM_TXN AVG_AMT_PER_CSH_WDL_TXN AVG_AMT_PER_CHQ_TXN
## Min. : 0 Min. : 0 Min. : 0
## 1st Qu.: 0 1st Qu.: 1266 1st Qu.: 0
## Median : 6000 Median :147095 Median : 8645
## Mean : 7409 Mean :242237 Mean : 25093
## 3rd Qu.:13500 3rd Qu.:385000 3rd Qu.: 28605
## Max. :25000 Max. :999640 Max. :537842
##
## AVG_AMT_PER_NET_TXN AVG_AMT_PER_MOB_TXN FLG_HAS_NOMINEE FLG_HAS_OLD_LOAN
## Min. : 0 Min. : 0 Min. :0.0000 Min. :0.0000
## 1st Qu.: 0 1st Qu.: 0 1st Qu.:1.0000 1st Qu.:0.0000
## Median : 0 Median : 0 Median :1.0000 Median :0.0000
## Mean :179059 Mean : 20304 Mean :0.9012 Mean :0.4929
## 3rd Qu.:257699 3rd Qu.: 0 3rd Qu.:1.0000 3rd Qu.:1.0000
## Max. :999854 Max. :199667 Max. :1.0000 Max. :1.0000
##
## random
## Min. :0.0000114
## 1st Qu.:0.2481866
## Median :0.5061214
## Mean :0.5019330
## 3rd Qu.:0.7535712
## Max. :0.9999471
## Checking for any missing values in the dataset
library(VIM)## Warning: package 'VIM' was built under R version 3.4.3## Loading required package: colorspace## Loading required package: grid## Loading required package: data.table## Warning: package 'data.table' was built under R version 3.4.2## VIM is ready to use.
## Since version 4.0.0 the GUI is in its own package VIMGUI.
##
## Please use the package to use the new (and old) GUI.## Suggestions and bug-reports can be submitted at: https://github.com/alexkowa/VIM/issues##
## Attaching package: 'VIM'## The following object is masked from 'package:datasets':
##
## sleeplibrary(mice)## Warning: package 'mice' was built under R version 3.4.3## Loading required package: latticeopar <- par(no.readonly = TRUE)
par(bg ="gray63", col="white", col.axis = "white", col.lab = "white", col.main = "white",col.sub = "white")
aggr(loan.data,prop = F,cex.axis = 0.4, numbers = T)
par(opar)This chart shows the number of missing values if there is any in the dataset.
Visualisation of all the independent variables are done
Correlation matrix plot
num.data <- subset(loan.data[-c(2,4,6,7,10,11,21,28,38,39)])
corr <- cor(num.data[,-1])
library(corrplot)## Warning: package 'corrplot' was built under R version 3.4.2## corrplot 0.84 loadedopar2 <- par(no.readonly = TRUE)
corrplot(corr,method = "circle",tl.cex = 0.5,tl.col = "black",number.cex = 0.55,bg = "grey14",
addgrid.col = "gray50", tl.offset = 2,col = colorRampPalette(c("blue1","ivory2","firebrick2"))(100))
Changing the format of few variables
loan.data$ACC_OP_DATE <- as.character(loan.data$ACC_OP_DATE)
loan.data$ACC_OP_DATE <- as.Date(loan.data$ACC_OP_DATE, format="%m/%d/%Y")
loan.data$FLG_HAS_CC <- as.factor(loan.data$FLG_HAS_CC)
loan.data$FLG_HAS_ANY_CHGS <- as.factor(loan.data$FLG_HAS_ANY_CHGS)
loan.data$FLG_HAS_NOMINEE <- as.factor(loan.data$FLG_HAS_NOMINEE)
loan.data$FLG_HAS_OLD_LOAN <- as.factor(loan.data$FLG_HAS_OLD_LOAN)Splitting the given dataset into development and holdout
library(caret)## Loading required package: ggplot2set.seed(1407)
split <- createDataPartition(loan.data$TARGET,p = 0.7,list = FALSE,times = 1)
View(split)
RFDF.dev <- loan.data[split,]
RFDF.holdout <- loan.data[-split,]Model development - Random Forest technique
library(randomForest)## Warning: package 'randomForest' was built under R version 3.4.3## randomForest 4.6-12## Type rfNews() to see new features/changes/bug fixes.##
## Attaching package: 'randomForest'## The following object is masked from 'package:ggplot2':
##
## marginRF <- randomForest(as.factor(TARGET) ~ ., data = RFDF.dev[,-1],
ntree=501, mtry = 7, nodesize = 140,
importance=TRUE)
print(RF)##
## Call:
## randomForest(formula = as.factor(TARGET) ~ ., data = RFDF.dev[, -1], ntree = 501, mtry = 7, nodesize = 140, importance = TRUE)
## Type of random forest: classification
## Number of trees: 501
## No. of variables tried at each split: 7
##
## OOB estimate of error rate: 12.09%
## Confusion matrix:
## 0 1 class.error
## 0 12283 0 0.0000000
## 1 1692 25 0.9854397From the above output, we find that the Out Of Bag (OOB) error rate is estimated as 12.06% which is the misclassification error rate of the model.
Plotting OOB error rates
plot(RF, main="")
legend("topright", c("OOB", "0", "1"), text.col=1:6, lty=1:3, col=1:3)
title(main="Error Rates Random Forest RFDF.dev")
On plotting the OOB error rate across the decision trees, it seems to indicate that after approximately after 21 number of decision trees, there is no significant reduction in the OOB error rate.
RF$err.rate## OOB 0 1
## [1,] 0.1274244 2.336861e-02 0.8887097
## [2,] 0.1238781 1.785235e-02 0.8998035
## [3,] 0.1239958 1.495428e-02 0.9121306
## [4,] 0.1246516 1.528699e-02 0.9145833
## [5,] 0.1253258 1.304191e-02 0.9333333
## [6,] 0.1243437 1.101092e-02 0.9377722
## [7,] 0.1230085 1.080260e-02 0.9307412
## [8,] 0.1228789 9.247688e-03 0.9400839
## [9,] 0.1217177 6.777980e-03 0.9454976
## [10,] 0.1224387 6.002302e-03 0.9564193
## [11,] 0.1214537 4.911591e-03 0.9554774
## [12,] 0.1210492 3.512785e-03 0.9614486
## [13,] 0.1210711 3.182894e-03 0.9638273
## [14,] 0.1207526 2.609476e-03 0.9650350
## [15,] 0.1209522 1.792553e-03 0.9731935
## [16,] 0.1207574 1.547483e-03 0.9732091
## [17,] 0.1208917 1.303038e-03 0.9761211
## [18,] 0.1207315 1.139972e-03 0.9761211
## [19,] 0.1204372 8.141996e-04 0.9761211
## [20,] 0.1210801 1.139879e-03 0.9790332
## [21,] 0.1207229 9.770396e-04 0.9772860
## [22,] 0.1206515 8.141996e-04 0.9778684
## [23,] 0.1207229 6.513597e-04 0.9796156
## [24,] 0.1206515 4.070998e-04 0.9807804
## [25,] 0.1205714 3.256533e-04 0.9807804
## [26,] 0.1207143 2.442400e-04 0.9825277
## [27,] 0.1207857 3.256533e-04 0.9825277
## [28,] 0.1207857 2.442400e-04 0.9831101
## [29,] 0.1209286 2.442400e-04 0.9842749
## [30,] 0.1208571 8.141334e-05 0.9848573
## [31,] 0.1207857 0.000000e+00 0.9848573
## [32,] 0.1208571 8.141334e-05 0.9848573
## [33,] 0.1206429 0.000000e+00 0.9836925
## [34,] 0.1207143 0.000000e+00 0.9842749
## [35,] 0.1208571 0.000000e+00 0.9854397
## [36,] 0.1210000 0.000000e+00 0.9866045
## [37,] 0.1211429 0.000000e+00 0.9877694
## [38,] 0.1212143 0.000000e+00 0.9883518
## [39,] 0.1210714 0.000000e+00 0.9871870
## [40,] 0.1210714 0.000000e+00 0.9871870
## [41,] 0.1212143 0.000000e+00 0.9883518
## [42,] 0.1214286 0.000000e+00 0.9900990
## [43,] 0.1214286 0.000000e+00 0.9900990
## [44,] 0.1213571 0.000000e+00 0.9895166
## [45,] 0.1211429 0.000000e+00 0.9877694
## [46,] 0.1212143 0.000000e+00 0.9883518
## [47,] 0.1212857 8.141334e-05 0.9883518
## [48,] 0.1211429 0.000000e+00 0.9877694
## [49,] 0.1213571 0.000000e+00 0.9895166
## [50,] 0.1212143 0.000000e+00 0.9883518
## [51,] 0.1211429 0.000000e+00 0.9877694
## [52,] 0.1207857 0.000000e+00 0.9848573
## [53,] 0.1207857 0.000000e+00 0.9848573
## [54,] 0.1207857 0.000000e+00 0.9848573
## [55,] 0.1208571 0.000000e+00 0.9854397
## [56,] 0.1207857 0.000000e+00 0.9848573
## [57,] 0.1210000 0.000000e+00 0.9866045
## [58,] 0.1211429 0.000000e+00 0.9877694
## [59,] 0.1210000 0.000000e+00 0.9866045
## [60,] 0.1209286 0.000000e+00 0.9860221
## [61,] 0.1209286 0.000000e+00 0.9860221
## [62,] 0.1210714 0.000000e+00 0.9871870
## [63,] 0.1207857 0.000000e+00 0.9848573
## [64,] 0.1209286 0.000000e+00 0.9860221
## [65,] 0.1209286 0.000000e+00 0.9860221
## [66,] 0.1207857 0.000000e+00 0.9848573
## [67,] 0.1208571 8.141334e-05 0.9848573
## [68,] 0.1207857 0.000000e+00 0.9848573
## [69,] 0.1207857 0.000000e+00 0.9848573
## [70,] 0.1208571 0.000000e+00 0.9854397
## [71,] 0.1207143 0.000000e+00 0.9842749
## [72,] 0.1210000 0.000000e+00 0.9866045
## [73,] 0.1210714 0.000000e+00 0.9871870
## [74,] 0.1210000 0.000000e+00 0.9866045
## [75,] 0.1209286 0.000000e+00 0.9860221
## [76,] 0.1211429 0.000000e+00 0.9877694
## [77,] 0.1210714 0.000000e+00 0.9871870
## [78,] 0.1210714 0.000000e+00 0.9871870
## [79,] 0.1210714 0.000000e+00 0.9871870
## [80,] 0.1210000 0.000000e+00 0.9866045
## [81,] 0.1210714 0.000000e+00 0.9871870
## [82,] 0.1210714 0.000000e+00 0.9871870
## [83,] 0.1208571 0.000000e+00 0.9854397
## [84,] 0.1210000 0.000000e+00 0.9866045
## [85,] 0.1210714 0.000000e+00 0.9871870
## [86,] 0.1210714 0.000000e+00 0.9871870
## [87,] 0.1210000 0.000000e+00 0.9866045
## [88,] 0.1211429 0.000000e+00 0.9877694
## [89,] 0.1209286 0.000000e+00 0.9860221
## [90,] 0.1210000 0.000000e+00 0.9866045
## [91,] 0.1210714 0.000000e+00 0.9871870
## [92,] 0.1211429 0.000000e+00 0.9877694
## [93,] 0.1212143 0.000000e+00 0.9883518
## [94,] 0.1210714 0.000000e+00 0.9871870
## [95,] 0.1212143 0.000000e+00 0.9883518
## [96,] 0.1210000 0.000000e+00 0.9866045
## [97,] 0.1208571 0.000000e+00 0.9854397
## [98,] 0.1207143 0.000000e+00 0.9842749
## [99,] 0.1209286 0.000000e+00 0.9860221
## [100,] 0.1210000 0.000000e+00 0.9866045
## [101,] 0.1209286 0.000000e+00 0.9860221
## [102,] 0.1211429 0.000000e+00 0.9877694
## [103,] 0.1211429 0.000000e+00 0.9877694
## [104,] 0.1209286 0.000000e+00 0.9860221
## [105,] 0.1210714 0.000000e+00 0.9871870
## [106,] 0.1209286 0.000000e+00 0.9860221
## [107,] 0.1210714 0.000000e+00 0.9871870
## [108,] 0.1210714 0.000000e+00 0.9871870
## [109,] 0.1209286 0.000000e+00 0.9860221
## [110,] 0.1210000 0.000000e+00 0.9866045
## [111,] 0.1210714 0.000000e+00 0.9871870
## [112,] 0.1209286 0.000000e+00 0.9860221
## [113,] 0.1208571 0.000000e+00 0.9854397
## [114,] 0.1209286 0.000000e+00 0.9860221
## [115,] 0.1208571 0.000000e+00 0.9854397
## [116,] 0.1210714 0.000000e+00 0.9871870
## [117,] 0.1210000 0.000000e+00 0.9866045
## [118,] 0.1210714 0.000000e+00 0.9871870
## [119,] 0.1210000 0.000000e+00 0.9866045
## [120,] 0.1210000 0.000000e+00 0.9866045
## [121,] 0.1210714 0.000000e+00 0.9871870
## [122,] 0.1210714 0.000000e+00 0.9871870
## [123,] 0.1210000 0.000000e+00 0.9866045
## [124,] 0.1210714 0.000000e+00 0.9871870
## [125,] 0.1210714 0.000000e+00 0.9871870
## [126,] 0.1210000 0.000000e+00 0.9866045
## [127,] 0.1210714 0.000000e+00 0.9871870
## [128,] 0.1211429 0.000000e+00 0.9877694
## [129,] 0.1209286 0.000000e+00 0.9860221
## [130,] 0.1210000 0.000000e+00 0.9866045
## [131,] 0.1210000 0.000000e+00 0.9866045
## [132,] 0.1211429 0.000000e+00 0.9877694
## [133,] 0.1210714 0.000000e+00 0.9871870
## [134,] 0.1210000 0.000000e+00 0.9866045
## [135,] 0.1211429 0.000000e+00 0.9877694
## [136,] 0.1211429 0.000000e+00 0.9877694
## [137,] 0.1211429 0.000000e+00 0.9877694
## [138,] 0.1211429 0.000000e+00 0.9877694
## [139,] 0.1212143 0.000000e+00 0.9883518
## [140,] 0.1212143 0.000000e+00 0.9883518
## [141,] 0.1210714 0.000000e+00 0.9871870
## [142,] 0.1209286 0.000000e+00 0.9860221
## [143,] 0.1209286 0.000000e+00 0.9860221
## [144,] 0.1210714 0.000000e+00 0.9871870
## [145,] 0.1210000 0.000000e+00 0.9866045
## [146,] 0.1210000 0.000000e+00 0.9866045
## [147,] 0.1208571 0.000000e+00 0.9854397
## [148,] 0.1209286 0.000000e+00 0.9860221
## [149,] 0.1208571 0.000000e+00 0.9854397
## [150,] 0.1209286 0.000000e+00 0.9860221
## [151,] 0.1209286 0.000000e+00 0.9860221
## [152,] 0.1210000 0.000000e+00 0.9866045
## [153,] 0.1210000 0.000000e+00 0.9866045
## [154,] 0.1210000 0.000000e+00 0.9866045
## [155,] 0.1210000 0.000000e+00 0.9866045
## [156,] 0.1209286 0.000000e+00 0.9860221
## [157,] 0.1209286 0.000000e+00 0.9860221
## [158,] 0.1210000 0.000000e+00 0.9866045
## [159,] 0.1210000 0.000000e+00 0.9866045
## [160,] 0.1210714 0.000000e+00 0.9871870
## [161,] 0.1209286 0.000000e+00 0.9860221
## [162,] 0.1210000 0.000000e+00 0.9866045
## [163,] 0.1208571 0.000000e+00 0.9854397
## [164,] 0.1209286 0.000000e+00 0.9860221
## [165,] 0.1208571 0.000000e+00 0.9854397
## [166,] 0.1207857 0.000000e+00 0.9848573
## [167,] 0.1207857 0.000000e+00 0.9848573
## [168,] 0.1207857 0.000000e+00 0.9848573
## [169,] 0.1208571 0.000000e+00 0.9854397
## [170,] 0.1208571 0.000000e+00 0.9854397
## [171,] 0.1208571 0.000000e+00 0.9854397
## [172,] 0.1210000 0.000000e+00 0.9866045
## [173,] 0.1211429 0.000000e+00 0.9877694
## [174,] 0.1210000 0.000000e+00 0.9866045
## [175,] 0.1210714 0.000000e+00 0.9871870
## [176,] 0.1210714 0.000000e+00 0.9871870
## [177,] 0.1210714 0.000000e+00 0.9871870
## [178,] 0.1208571 0.000000e+00 0.9854397
## [179,] 0.1210000 0.000000e+00 0.9866045
## [180,] 0.1210000 0.000000e+00 0.9866045
## [181,] 0.1208571 0.000000e+00 0.9854397
## [182,] 0.1208571 0.000000e+00 0.9854397
## [183,] 0.1210000 0.000000e+00 0.9866045
## [184,] 0.1209286 0.000000e+00 0.9860221
## [185,] 0.1210714 0.000000e+00 0.9871870
## [186,] 0.1209286 0.000000e+00 0.9860221
## [187,] 0.1209286 0.000000e+00 0.9860221
## [188,] 0.1208571 0.000000e+00 0.9854397
## [189,] 0.1208571 0.000000e+00 0.9854397
## [190,] 0.1209286 0.000000e+00 0.9860221
## [191,] 0.1209286 0.000000e+00 0.9860221
## [192,] 0.1209286 0.000000e+00 0.9860221
## [193,] 0.1210000 0.000000e+00 0.9866045
## [194,] 0.1210000 0.000000e+00 0.9866045
## [195,] 0.1209286 0.000000e+00 0.9860221
## [196,] 0.1210000 0.000000e+00 0.9866045
## [197,] 0.1210000 0.000000e+00 0.9866045
## [198,] 0.1210000 0.000000e+00 0.9866045
## [199,] 0.1210000 0.000000e+00 0.9866045
## [200,] 0.1210714 0.000000e+00 0.9871870
## [201,] 0.1210714 0.000000e+00 0.9871870
## [202,] 0.1210000 0.000000e+00 0.9866045
## [203,] 0.1209286 0.000000e+00 0.9860221
## [204,] 0.1211429 0.000000e+00 0.9877694
## [205,] 0.1210000 0.000000e+00 0.9866045
## [206,] 0.1210714 0.000000e+00 0.9871870
## [207,] 0.1210714 0.000000e+00 0.9871870
## [208,] 0.1209286 0.000000e+00 0.9860221
## [209,] 0.1209286 0.000000e+00 0.9860221
## [210,] 0.1210000 0.000000e+00 0.9866045
## [211,] 0.1210000 0.000000e+00 0.9866045
## [212,] 0.1210000 0.000000e+00 0.9866045
## [213,] 0.1209286 0.000000e+00 0.9860221
## [214,] 0.1210000 0.000000e+00 0.9866045
## [215,] 0.1209286 0.000000e+00 0.9860221
## [216,] 0.1208571 0.000000e+00 0.9854397
## [217,] 0.1207857 0.000000e+00 0.9848573
## [218,] 0.1209286 0.000000e+00 0.9860221
## [219,] 0.1208571 0.000000e+00 0.9854397
## [220,] 0.1208571 0.000000e+00 0.9854397
## [221,] 0.1208571 0.000000e+00 0.9854397
## [222,] 0.1209286 0.000000e+00 0.9860221
## [223,] 0.1208571 0.000000e+00 0.9854397
## [224,] 0.1207857 0.000000e+00 0.9848573
## [225,] 0.1207857 0.000000e+00 0.9848573
## [226,] 0.1207857 0.000000e+00 0.9848573
## [227,] 0.1208571 0.000000e+00 0.9854397
## [228,] 0.1208571 0.000000e+00 0.9854397
## [229,] 0.1208571 0.000000e+00 0.9854397
## [230,] 0.1208571 0.000000e+00 0.9854397
## [231,] 0.1208571 0.000000e+00 0.9854397
## [232,] 0.1208571 0.000000e+00 0.9854397
## [233,] 0.1208571 0.000000e+00 0.9854397
## [234,] 0.1207857 0.000000e+00 0.9848573
## [235,] 0.1208571 0.000000e+00 0.9854397
## [236,] 0.1208571 0.000000e+00 0.9854397
## [237,] 0.1208571 0.000000e+00 0.9854397
## [238,] 0.1209286 0.000000e+00 0.9860221
## [239,] 0.1208571 0.000000e+00 0.9854397
## [240,] 0.1209286 0.000000e+00 0.9860221
## [241,] 0.1209286 0.000000e+00 0.9860221
## [242,] 0.1209286 0.000000e+00 0.9860221
## [243,] 0.1209286 0.000000e+00 0.9860221
## [244,] 0.1209286 0.000000e+00 0.9860221
## [245,] 0.1208571 0.000000e+00 0.9854397
## [246,] 0.1208571 0.000000e+00 0.9854397
## [247,] 0.1209286 0.000000e+00 0.9860221
## [248,] 0.1210000 0.000000e+00 0.9866045
## [249,] 0.1209286 0.000000e+00 0.9860221
## [250,] 0.1210000 0.000000e+00 0.9866045
## [251,] 0.1209286 0.000000e+00 0.9860221
## [252,] 0.1209286 0.000000e+00 0.9860221
## [253,] 0.1209286 0.000000e+00 0.9860221
## [254,] 0.1209286 0.000000e+00 0.9860221
## [255,] 0.1209286 0.000000e+00 0.9860221
## [256,] 0.1210000 0.000000e+00 0.9866045
## [257,] 0.1209286 0.000000e+00 0.9860221
## [258,] 0.1209286 0.000000e+00 0.9860221
## [259,] 0.1209286 0.000000e+00 0.9860221
## [260,] 0.1209286 0.000000e+00 0.9860221
## [261,] 0.1209286 0.000000e+00 0.9860221
## [262,] 0.1209286 0.000000e+00 0.9860221
## [263,] 0.1209286 0.000000e+00 0.9860221
## [264,] 0.1209286 0.000000e+00 0.9860221
## [265,] 0.1209286 0.000000e+00 0.9860221
## [266,] 0.1209286 0.000000e+00 0.9860221
## [267,] 0.1209286 0.000000e+00 0.9860221
## [268,] 0.1209286 0.000000e+00 0.9860221
## [269,] 0.1209286 0.000000e+00 0.9860221
## [270,] 0.1209286 0.000000e+00 0.9860221
## [271,] 0.1209286 0.000000e+00 0.9860221
## [272,] 0.1209286 0.000000e+00 0.9860221
## [273,] 0.1208571 0.000000e+00 0.9854397
## [274,] 0.1209286 0.000000e+00 0.9860221
## [275,] 0.1209286 0.000000e+00 0.9860221
## [276,] 0.1209286 0.000000e+00 0.9860221
## [277,] 0.1209286 0.000000e+00 0.9860221
## [278,] 0.1209286 0.000000e+00 0.9860221
## [279,] 0.1209286 0.000000e+00 0.9860221
## [280,] 0.1209286 0.000000e+00 0.9860221
## [281,] 0.1209286 0.000000e+00 0.9860221
## [282,] 0.1209286 0.000000e+00 0.9860221
## [283,] 0.1209286 0.000000e+00 0.9860221
## [284,] 0.1209286 0.000000e+00 0.9860221
## [285,] 0.1209286 0.000000e+00 0.9860221
## [286,] 0.1209286 0.000000e+00 0.9860221
## [287,] 0.1210000 0.000000e+00 0.9866045
## [288,] 0.1209286 0.000000e+00 0.9860221
## [289,] 0.1210000 0.000000e+00 0.9866045
## [290,] 0.1210000 0.000000e+00 0.9866045
## [291,] 0.1210000 0.000000e+00 0.9866045
## [292,] 0.1209286 0.000000e+00 0.9860221
## [293,] 0.1209286 0.000000e+00 0.9860221
## [294,] 0.1210000 0.000000e+00 0.9866045
## [295,] 0.1209286 0.000000e+00 0.9860221
## [296,] 0.1208571 0.000000e+00 0.9854397
## [297,] 0.1208571 0.000000e+00 0.9854397
## [298,] 0.1208571 0.000000e+00 0.9854397
## [299,] 0.1208571 0.000000e+00 0.9854397
## [300,] 0.1208571 0.000000e+00 0.9854397
## [301,] 0.1208571 0.000000e+00 0.9854397
## [302,] 0.1208571 0.000000e+00 0.9854397
## [303,] 0.1208571 0.000000e+00 0.9854397
## [304,] 0.1208571 0.000000e+00 0.9854397
## [305,] 0.1207857 0.000000e+00 0.9848573
## [306,] 0.1207857 0.000000e+00 0.9848573
## [307,] 0.1207857 0.000000e+00 0.9848573
## [308,] 0.1208571 0.000000e+00 0.9854397
## [309,] 0.1207857 0.000000e+00 0.9848573
## [310,] 0.1207143 0.000000e+00 0.9842749
## [311,] 0.1207143 0.000000e+00 0.9842749
## [312,] 0.1207143 0.000000e+00 0.9842749
## [313,] 0.1207857 0.000000e+00 0.9848573
## [314,] 0.1207143 0.000000e+00 0.9842749
## [315,] 0.1207857 0.000000e+00 0.9848573
## [316,] 0.1207143 0.000000e+00 0.9842749
## [317,] 0.1207857 0.000000e+00 0.9848573
## [318,] 0.1207143 0.000000e+00 0.9842749
## [319,] 0.1207143 0.000000e+00 0.9842749
## [320,] 0.1206429 0.000000e+00 0.9836925
## [321,] 0.1206429 0.000000e+00 0.9836925
## [322,] 0.1207857 0.000000e+00 0.9848573
## [323,] 0.1207143 0.000000e+00 0.9842749
## [324,] 0.1207857 0.000000e+00 0.9848573
## [325,] 0.1207857 0.000000e+00 0.9848573
## [326,] 0.1207143 0.000000e+00 0.9842749
## [327,] 0.1207857 0.000000e+00 0.9848573
## [328,] 0.1207857 0.000000e+00 0.9848573
## [329,] 0.1207857 0.000000e+00 0.9848573
## [330,] 0.1207857 0.000000e+00 0.9848573
## [331,] 0.1207857 0.000000e+00 0.9848573
## [332,] 0.1207857 0.000000e+00 0.9848573
## [333,] 0.1207857 0.000000e+00 0.9848573
## [334,] 0.1207143 0.000000e+00 0.9842749
## [335,] 0.1207143 0.000000e+00 0.9842749
## [336,] 0.1207143 0.000000e+00 0.9842749
## [337,] 0.1207143 0.000000e+00 0.9842749
## [338,] 0.1207143 0.000000e+00 0.9842749
## [339,] 0.1207857 0.000000e+00 0.9848573
## [340,] 0.1208571 0.000000e+00 0.9854397
## [341,] 0.1207143 0.000000e+00 0.9842749
## [342,] 0.1207143 0.000000e+00 0.9842749
## [343,] 0.1207143 0.000000e+00 0.9842749
## [344,] 0.1207143 0.000000e+00 0.9842749
## [345,] 0.1207143 0.000000e+00 0.9842749
## [346,] 0.1207143 0.000000e+00 0.9842749
## [347,] 0.1207143 0.000000e+00 0.9842749
## [348,] 0.1207857 0.000000e+00 0.9848573
## [349,] 0.1207143 0.000000e+00 0.9842749
## [350,] 0.1207857 0.000000e+00 0.9848573
## [351,] 0.1207857 0.000000e+00 0.9848573
## [352,] 0.1207857 0.000000e+00 0.9848573
## [353,] 0.1207857 0.000000e+00 0.9848573
## [354,] 0.1207143 0.000000e+00 0.9842749
## [355,] 0.1207857 0.000000e+00 0.9848573
## [356,] 0.1207143 0.000000e+00 0.9842749
## [357,] 0.1207857 0.000000e+00 0.9848573
## [358,] 0.1207857 0.000000e+00 0.9848573
## [359,] 0.1207857 0.000000e+00 0.9848573
## [360,] 0.1207857 0.000000e+00 0.9848573
## [361,] 0.1207857 0.000000e+00 0.9848573
## [362,] 0.1207857 0.000000e+00 0.9848573
## [363,] 0.1207857 0.000000e+00 0.9848573
## [364,] 0.1207857 0.000000e+00 0.9848573
## [365,] 0.1207857 0.000000e+00 0.9848573
## [366,] 0.1207857 0.000000e+00 0.9848573
## [367,] 0.1207857 0.000000e+00 0.9848573
## [368,] 0.1207857 0.000000e+00 0.9848573
## [369,] 0.1207857 0.000000e+00 0.9848573
## [370,] 0.1207857 0.000000e+00 0.9848573
## [371,] 0.1207857 0.000000e+00 0.9848573
## [372,] 0.1207857 0.000000e+00 0.9848573
## [373,] 0.1207857 0.000000e+00 0.9848573
## [374,] 0.1207857 0.000000e+00 0.9848573
## [375,] 0.1207857 0.000000e+00 0.9848573
## [376,] 0.1207857 0.000000e+00 0.9848573
## [377,] 0.1207857 0.000000e+00 0.9848573
## [378,] 0.1207143 0.000000e+00 0.9842749
## [379,] 0.1207143 0.000000e+00 0.9842749
## [380,] 0.1207143 0.000000e+00 0.9842749
## [381,] 0.1207143 0.000000e+00 0.9842749
## [382,] 0.1207143 0.000000e+00 0.9842749
## [383,] 0.1207143 0.000000e+00 0.9842749
## [384,] 0.1207143 0.000000e+00 0.9842749
## [385,] 0.1207143 0.000000e+00 0.9842749
## [386,] 0.1207857 0.000000e+00 0.9848573
## [387,] 0.1207857 0.000000e+00 0.9848573
## [388,] 0.1207857 0.000000e+00 0.9848573
## [389,] 0.1207857 0.000000e+00 0.9848573
## [390,] 0.1207857 0.000000e+00 0.9848573
## [391,] 0.1207857 0.000000e+00 0.9848573
## [392,] 0.1207857 0.000000e+00 0.9848573
## [393,] 0.1207857 0.000000e+00 0.9848573
## [394,] 0.1207857 0.000000e+00 0.9848573
## [395,] 0.1207857 0.000000e+00 0.9848573
## [396,] 0.1207857 0.000000e+00 0.9848573
## [397,] 0.1207143 0.000000e+00 0.9842749
## [398,] 0.1207857 0.000000e+00 0.9848573
## [399,] 0.1207857 0.000000e+00 0.9848573
## [400,] 0.1207857 0.000000e+00 0.9848573
## [401,] 0.1207857 0.000000e+00 0.9848573
## [402,] 0.1207857 0.000000e+00 0.9848573
## [403,] 0.1207857 0.000000e+00 0.9848573
## [404,] 0.1207857 0.000000e+00 0.9848573
## [405,] 0.1207857 0.000000e+00 0.9848573
## [406,] 0.1207857 0.000000e+00 0.9848573
## [407,] 0.1207857 0.000000e+00 0.9848573
## [408,] 0.1207857 0.000000e+00 0.9848573
## [409,] 0.1207857 0.000000e+00 0.9848573
## [410,] 0.1207857 0.000000e+00 0.9848573
## [411,] 0.1207857 0.000000e+00 0.9848573
## [412,] 0.1207857 0.000000e+00 0.9848573
## [413,] 0.1207857 0.000000e+00 0.9848573
## [414,] 0.1207857 0.000000e+00 0.9848573
## [415,] 0.1207857 0.000000e+00 0.9848573
## [416,] 0.1207857 0.000000e+00 0.9848573
## [417,] 0.1207857 0.000000e+00 0.9848573
## [418,] 0.1207857 0.000000e+00 0.9848573
## [419,] 0.1207857 0.000000e+00 0.9848573
## [420,] 0.1207857 0.000000e+00 0.9848573
## [421,] 0.1207857 0.000000e+00 0.9848573
## [422,] 0.1207857 0.000000e+00 0.9848573
## [423,] 0.1207857 0.000000e+00 0.9848573
## [424,] 0.1208571 0.000000e+00 0.9854397
## [425,] 0.1208571 0.000000e+00 0.9854397
## [426,] 0.1209286 0.000000e+00 0.9860221
## [427,] 0.1208571 0.000000e+00 0.9854397
## [428,] 0.1208571 0.000000e+00 0.9854397
## [429,] 0.1209286 0.000000e+00 0.9860221
## [430,] 0.1209286 0.000000e+00 0.9860221
## [431,] 0.1209286 0.000000e+00 0.9860221
## [432,] 0.1209286 0.000000e+00 0.9860221
## [433,] 0.1209286 0.000000e+00 0.9860221
## [434,] 0.1209286 0.000000e+00 0.9860221
## [435,] 0.1209286 0.000000e+00 0.9860221
## [436,] 0.1209286 0.000000e+00 0.9860221
## [437,] 0.1209286 0.000000e+00 0.9860221
## [438,] 0.1209286 0.000000e+00 0.9860221
## [439,] 0.1208571 0.000000e+00 0.9854397
## [440,] 0.1208571 0.000000e+00 0.9854397
## [441,] 0.1207857 0.000000e+00 0.9848573
## [442,] 0.1209286 0.000000e+00 0.9860221
## [443,] 0.1208571 0.000000e+00 0.9854397
## [444,] 0.1208571 0.000000e+00 0.9854397
## [445,] 0.1208571 0.000000e+00 0.9854397
## [446,] 0.1208571 0.000000e+00 0.9854397
## [447,] 0.1209286 0.000000e+00 0.9860221
## [448,] 0.1209286 0.000000e+00 0.9860221
## [449,] 0.1209286 0.000000e+00 0.9860221
## [450,] 0.1209286 0.000000e+00 0.9860221
## [451,] 0.1209286 0.000000e+00 0.9860221
## [452,] 0.1208571 0.000000e+00 0.9854397
## [453,] 0.1207857 0.000000e+00 0.9848573
## [454,] 0.1209286 0.000000e+00 0.9860221
## [455,] 0.1207857 0.000000e+00 0.9848573
## [456,] 0.1208571 0.000000e+00 0.9854397
## [457,] 0.1208571 0.000000e+00 0.9854397
## [458,] 0.1208571 0.000000e+00 0.9854397
## [459,] 0.1208571 0.000000e+00 0.9854397
## [460,] 0.1208571 0.000000e+00 0.9854397
## [461,] 0.1208571 0.000000e+00 0.9854397
## [462,] 0.1210000 0.000000e+00 0.9866045
## [463,] 0.1210000 0.000000e+00 0.9866045
## [464,] 0.1209286 0.000000e+00 0.9860221
## [465,] 0.1208571 0.000000e+00 0.9854397
## [466,] 0.1208571 0.000000e+00 0.9854397
## [467,] 0.1208571 0.000000e+00 0.9854397
## [468,] 0.1210000 0.000000e+00 0.9866045
## [469,] 0.1208571 0.000000e+00 0.9854397
## [470,] 0.1210000 0.000000e+00 0.9866045
## [471,] 0.1208571 0.000000e+00 0.9854397
## [472,] 0.1209286 0.000000e+00 0.9860221
## [473,] 0.1209286 0.000000e+00 0.9860221
## [474,] 0.1209286 0.000000e+00 0.9860221
## [475,] 0.1209286 0.000000e+00 0.9860221
## [476,] 0.1208571 0.000000e+00 0.9854397
## [477,] 0.1208571 0.000000e+00 0.9854397
## [478,] 0.1208571 0.000000e+00 0.9854397
## [479,] 0.1207857 0.000000e+00 0.9848573
## [480,] 0.1208571 0.000000e+00 0.9854397
## [481,] 0.1207857 0.000000e+00 0.9848573
## [482,] 0.1207857 0.000000e+00 0.9848573
## [483,] 0.1208571 0.000000e+00 0.9854397
## [484,] 0.1207143 0.000000e+00 0.9842749
## [485,] 0.1207857 0.000000e+00 0.9848573
## [486,] 0.1207143 0.000000e+00 0.9842749
## [487,] 0.1207143 0.000000e+00 0.9842749
## [488,] 0.1207143 0.000000e+00 0.9842749
## [489,] 0.1207857 0.000000e+00 0.9848573
## [490,] 0.1207857 0.000000e+00 0.9848573
## [491,] 0.1207857 0.000000e+00 0.9848573
## [492,] 0.1207857 0.000000e+00 0.9848573
## [493,] 0.1207857 0.000000e+00 0.9848573
## [494,] 0.1207857 0.000000e+00 0.9848573
## [495,] 0.1207143 0.000000e+00 0.9842749
## [496,] 0.1207143 0.000000e+00 0.9842749
## [497,] 0.1207143 0.000000e+00 0.9842749
## [498,] 0.1207857 0.000000e+00 0.9848573
## [499,] 0.1207857 0.000000e+00 0.9848573
## [500,] 0.1207857 0.000000e+00 0.9848573
## [501,] 0.1208571 0.000000e+00 0.9854397Tuning Random Forest model
tRF <- tuneRF(x = RFDF.dev[,-c(1,2)],
y=as.factor(RFDF.dev$TARGET),
mtryStart = 7,
ntreeTry=21,
stepFactor = 1.5,
improve = 0.001,
trace = TRUE,
plot = TRUE,
doBest = TRUE,
nodesize = 140,
importance= TRUE
)## mtry = 7 OOB error = 12.02%
## Searching left ...
## mtry = 5 OOB error = 11.91%
## 0.009648332 0.001
## mtry = 4 OOB error = 12.16%
## -0.0209958 0.001
## Searching right ...
## mtry = 10 OOB error = 11.97%
## -0.004870817 0.001
print(tRF)##
## Call:
## randomForest(x = x, y = y, mtry = res[which.min(res[, 2]), 1], nodesize = 140, importance = TRUE)
## Type of random forest: classification
## Number of trees: 500
## No. of variables tried at each split: 5
##
## OOB estimate of error rate: 12.16%
## Confusion matrix:
## 0 1 class.error
## 0 12283 0 0.0000000
## 1 1703 14 0.9918462Identifying the significant varibales in the dataset
tRF$importance## 0 1 MeanDecreaseAccuracy
## AGE 3.425897e-04 1.944746e-03 5.403062e-04
## GENDER 5.602546e-04 2.159199e-03 7.560659e-04
## BALANCE 7.762846e-04 1.143337e-02 2.079991e-03
## OCCUPATION 1.377326e-03 1.095613e-02 2.555519e-03
## AGE_BKT 4.583121e-04 3.154720e-03 7.891677e-04
## SCR 8.933981e-04 1.025355e-02 2.042070e-03
## HOLDING_PERIOD 4.563459e-03 6.315845e-03 4.774016e-03
## ACC_TYPE 1.778926e-03 -1.320389e-03 1.400183e-03
## ACC_OP_DATE 2.885858e-03 5.268199e-04 2.594024e-03
## LEN_OF_RLTN_IN_MNTH 2.772508e-03 -2.309647e-04 2.400984e-03
## NO_OF_L_CR_TXNS 1.470998e-02 -3.876875e-03 1.242617e-02
## NO_OF_L_DR_TXNS 1.837587e-02 -6.761351e-03 1.528969e-02
## TOT_NO_OF_L_TXNS 1.664199e-02 -4.842343e-03 1.400249e-02
## NO_OF_BR_CSH_WDL_DR_TXNS 1.010543e-03 1.661191e-03 1.090779e-03
## NO_OF_ATM_DR_TXNS 1.947562e-02 -1.063444e-02 1.578222e-02
## NO_OF_NET_DR_TXNS 3.074151e-03 1.609162e-04 2.716221e-03
## NO_OF_MOB_DR_TXNS 1.842988e-03 -1.040690e-03 1.489555e-03
## NO_OF_CHQ_DR_TXNS 5.474617e-03 -6.611069e-04 4.718458e-03
## FLG_HAS_CC 8.107237e-04 8.042513e-03 1.698962e-03
## AMT_ATM_DR 9.628617e-03 -4.360724e-03 7.900483e-03
## AMT_BR_CSH_WDL_DR 3.222519e-03 1.130201e-03 2.967169e-03
## AMT_CHQ_DR 8.159171e-03 -2.767488e-03 6.802306e-03
## AMT_NET_DR 3.549619e-03 -2.216997e-04 3.083763e-03
## AMT_MOB_DR 3.034428e-03 -1.390956e-03 2.487666e-03
## AMT_L_DR 7.265713e-03 -1.418221e-03 6.196285e-03
## FLG_HAS_ANY_CHGS 7.499825e-05 2.774578e-04 9.985385e-05
## AMT_OTH_BK_ATM_USG_CHGS -3.090926e-06 4.233552e-05 2.328998e-06
## AMT_MIN_BAL_NMC_CHGS 1.675474e-05 7.234532e-05 2.370808e-05
## NO_OF_IW_CHQ_BNC_TXNS 3.280957e-05 1.229900e-04 4.384442e-05
## NO_OF_OW_CHQ_BNC_TXNS 5.618289e-05 1.498884e-04 6.759989e-05
## AVG_AMT_PER_ATM_TXN 1.027024e-02 -4.515455e-03 8.470218e-03
## AVG_AMT_PER_CSH_WDL_TXN 3.034307e-03 2.288898e-04 2.690810e-03
## AVG_AMT_PER_CHQ_TXN 7.464752e-03 -3.159354e-03 6.148180e-03
## AVG_AMT_PER_NET_TXN 3.569669e-03 -1.368415e-03 2.959954e-03
## AVG_AMT_PER_MOB_TXN 2.961331e-03 -3.690877e-04 2.554043e-03
## FLG_HAS_NOMINEE 7.129803e-06 2.545406e-05 9.351678e-06
## FLG_HAS_OLD_LOAN 4.202514e-05 1.788906e-04 5.857682e-05
## random -6.106719e-06 -5.028254e-05 -1.121834e-05
## MeanDecreaseGini
## AGE 12.0126492
## GENDER 9.1978994
## BALANCE 36.6731768
## OCCUPATION 31.7748278
## AGE_BKT 17.5272337
## SCR 40.7778470
## HOLDING_PERIOD 35.5695135
## ACC_TYPE 3.3228753
## ACC_OP_DATE 19.3610098
## LEN_OF_RLTN_IN_MNTH 15.3846344
## NO_OF_L_CR_TXNS 36.1436359
## NO_OF_L_DR_TXNS 18.0994543
## TOT_NO_OF_L_TXNS 35.8331748
## NO_OF_BR_CSH_WDL_DR_TXNS 7.6040447
## NO_OF_ATM_DR_TXNS 8.3900361
## NO_OF_NET_DR_TXNS 7.7234390
## NO_OF_MOB_DR_TXNS 2.1392640
## NO_OF_CHQ_DR_TXNS 8.0845459
## FLG_HAS_CC 18.9783743
## AMT_ATM_DR 17.0381130
## AMT_BR_CSH_WDL_DR 15.9813258
## AMT_CHQ_DR 16.5204370
## AMT_NET_DR 11.7727981
## AMT_MOB_DR 8.6954668
## AMT_L_DR 23.0445017
## FLG_HAS_ANY_CHGS 2.1862448
## AMT_OTH_BK_ATM_USG_CHGS 0.3696771
## AMT_MIN_BAL_NMC_CHGS 0.5864845
## NO_OF_IW_CHQ_BNC_TXNS 1.5312316
## NO_OF_OW_CHQ_BNC_TXNS 1.7435830
## AVG_AMT_PER_ATM_TXN 18.8887459
## AVG_AMT_PER_CSH_WDL_TXN 15.6072288
## AVG_AMT_PER_CHQ_TXN 15.0963810
## AVG_AMT_PER_NET_TXN 10.9166934
## AVG_AMT_PER_MOB_TXN 10.4919516
## FLG_HAS_NOMINEE 0.7164978
## FLG_HAS_OLD_LOAN 1.2479892
## random 7.8303132Plot of significant variables based on MeanDecreaseAccuracy and MeanDecreaseGini
varImpPlot(tRF,
sort = T,
main="Variable Importance",
n.var=38)
Scoring syntax
RFDF.dev$predict.class <- predict(tRF, RFDF.dev, type="class")
RFDF.dev$predict.score <- predict(tRF, RFDF.dev, type="prob")
head(RFDF.dev)## CUST_ID TARGET AGE GENDER BALANCE OCCUPATION AGE_BKT SCR
## 2 C6877 0 47 M 287489.04 SAL 46-50 324
## 3 C19922 0 40 M 18216.88 SELF-EMP 36-40 603
## 4 C8183 0 53 M 71720.48 SAL >50 196
## 5 C12123 0 36 M 1671622.89 PROF 36-40 167
## 6 C257 0 42 F 521685.69 PROF 41-45 493
## 7 C8300 0 30 M 204458.60 PROF 26-30 479
## HOLDING_PERIOD ACC_TYPE ACC_OP_DATE LEN_OF_RLTN_IN_MNTH NO_OF_L_CR_TXNS
## 2 28 SA 2008-10-11 104 8
## 3 2 SA 2012-04-26 61 10
## 4 13 CA 2008-07-04 107 36
## 5 24 SA 2001-12-29 185 20
## 6 26 SA 2001-06-07 192 5
## 7 14 SA 2002-08-25 177 6
## NO_OF_L_DR_TXNS TOT_NO_OF_L_TXNS NO_OF_BR_CSH_WDL_DR_TXNS
## 2 2 10 0
## 3 5 15 1
## 4 14 50 4
## 5 1 21 1
## 6 2 7 1
## 7 6 12 0
## NO_OF_ATM_DR_TXNS NO_OF_NET_DR_TXNS NO_OF_MOB_DR_TXNS NO_OF_CHQ_DR_TXNS
## 2 1 1 0 0
## 3 1 1 0 2
## 4 2 3 1 4
## 5 0 0 0 0
## 6 1 0 0 0
## 7 1 1 0 4
## FLG_HAS_CC AMT_ATM_DR AMT_BR_CSH_WDL_DR AMT_CHQ_DR AMT_NET_DR AMT_MOB_DR
## 2 0 6600 0 0 799813 0
## 3 0 11200 561120 49320 997570 0
## 4 0 26100 673590 60780 741506 71388
## 5 0 0 808480 0 0 0
## 6 1 18500 379310 0 0 0
## 7 0 6200 0 10580 770065 0
## AMT_L_DR FLG_HAS_ANY_CHGS AMT_OTH_BK_ATM_USG_CHGS AMT_MIN_BAL_NMC_CHGS
## 2 806413 1 0 0
## 3 1619210 1 0 0
## 4 1573364 0 0 0
## 5 808480 0 0 0
## 6 397810 0 0 0
## 7 786845 1 0 0
## NO_OF_IW_CHQ_BNC_TXNS NO_OF_OW_CHQ_BNC_TXNS AVG_AMT_PER_ATM_TXN
## 2 0 0 6600
## 3 0 1 11200
## 4 0 0 13050
## 5 0 0 0
## 6 0 0 18500
## 7 0 0 6200
## AVG_AMT_PER_CSH_WDL_TXN AVG_AMT_PER_CHQ_TXN AVG_AMT_PER_NET_TXN
## 2 0.0 0 799813.0
## 3 561120.0 24660 997570.0
## 4 168397.5 15195 247168.7
## 5 808480.0 0 0.0
## 6 379310.0 0 0.0
## 7 0.0 2645 770065.0
## AVG_AMT_PER_MOB_TXN FLG_HAS_NOMINEE FLG_HAS_OLD_LOAN random
## 2 0 1 0 0.000111373
## 3 0 1 1 0.000119954
## 4 71388 1 0 0.000136825
## 5 0 1 0 0.000173976
## 6 0 1 1 0.000405840
## 7 0 0 1 0.000499109
## predict.class predict.score.0 predict.score.1
## 2 0 0.994 0.006
## 3 0 0.922 0.078
## 4 0 0.994 0.006
## 5 0 0.986 0.014
## 6 0 0.994 0.006
## 7 0 0.994 0.006Model Performance Measures
Rank ordering
decile <- function(x){
deciles <- vector(length=10)
for (i in seq(0.1,1,.1)){
deciles[i*10] <- quantile(x, i, na.rm=T)
}
return (
ifelse(x<deciles[1], 1,
ifelse(x<deciles[2], 2,
ifelse(x<deciles[3], 3,
ifelse(x<deciles[4], 4,
ifelse(x<deciles[5], 5,
ifelse(x<deciles[6], 6,
ifelse(x<deciles[7], 7,
ifelse(x<deciles[8], 8,
ifelse(x<deciles[9], 9, 10
))))))))))
}
RFDF.dev$deciles <- decile(RFDF.dev$predict.score[,2])library(data.table)
tmp_DT = data.table(RFDF.dev)
rank <- tmp_DT[, list(
cnt = length(TARGET),
cnt_resp = sum(TARGET),
cnt_non_resp = sum(TARGET == 0)),
by=deciles][order(-deciles)]
rank$rrate <- round (rank$cnt_resp / rank$cnt,2);
rank$cum_resp <- cumsum(rank$cnt_resp)
rank$cum_non_resp <- cumsum(rank$cnt_non_resp)
rank$cum_rel_resp <- round(rank$cum_resp / sum(rank$cnt_resp),2);
rank$cum_rel_non_resp <- round(rank$cum_non_resp / sum(rank$cnt_non_resp),2);
rank$ks <- abs(rank$cum_rel_resp - rank$cum_rel_non_resp);library(scales)
rank$rrate <- percent(rank$rrate)
rank$cum_rel_resp <- percent(rank$cum_rel_resp)
rank$cum_rel_non_resp <- percent(rank$cum_rel_non_resp)
rank## deciles cnt cnt_resp cnt_non_resp rrate cum_resp cum_non_resp
## 1: 10 1410 986 424 70% 986 424
## 2: 9 1478 480 998 32% 1466 1422
## 3: 8 1423 151 1272 11% 1617 2694
## 4: 7 1366 65 1301 5% 1682 3995
## 5: 6 1799 17 1782 1% 1699 5777
## 6: 5 1372 10 1362 1% 1709 7139
## 7: 4 2183 4 2179 0% 1713 9318
## 8: 3 2969 4 2965 0% 1717 12283
## cum_rel_resp cum_rel_non_resp ks
## 1: 57% 3% 0.54
## 2: 85% 12% 0.73
## 3: 94% 22% 0.72
## 4: 98% 33% 0.65
## 5: 99% 47% 0.52
## 6: 100% 58% 0.42
## 7: 100% 76% 0.24
## 8: 100% 100% 0.00ROC Curve
library(ROCR)## Warning: package 'ROCR' was built under R version 3.4.2## Loading required package: gplots## Warning: package 'gplots' was built under R version 3.4.2##
## Attaching package: 'gplots'## The following object is masked from 'package:stats':
##
## lowesspred <- prediction(RFDF.dev$predict.score[,2],RFDF.dev$TARGET)
perf <- performance(pred, "tpr", "fpr")
plot(perf)
KS Value
KS <- max(attr(perf, 'y.values')[[1]]-attr(perf, 'x.values')[[1]])
KS## [1] 0.7452088Gini coefficient
library(ineq)
gini = ineq(RFDF.dev$predict.score[,2], type="Gini")
gini## [1] 0.7583828Validation of model using holdout sample
Rank ordering
RFDF.holdout$predict.class <- predict(tRF, RFDF.holdout, type="class")
RFDF.holdout$predict.score <- predict(tRF, RFDF.holdout, type="prob")
RFDF.holdout$deciles <- decile(RFDF.holdout$predict.score[,2])
tmp_DT = data.table(RFDF.holdout)
h_rank <- tmp_DT[, list(
cnt = length(TARGET),
cnt_resp = sum(TARGET),
cnt_non_resp = sum(TARGET == 0)) ,
by=deciles][order(-deciles)]
h_rank$rrate <- round (h_rank$cnt_resp / h_rank$cnt,2);
h_rank$cum_resp <- cumsum(h_rank$cnt_resp)
h_rank$cum_non_resp <- cumsum(h_rank$cnt_non_resp)
h_rank$cum_rel_resp <- round(h_rank$cum_resp / sum(h_rank$cnt_resp),2);
h_rank$cum_rel_non_resp <- round(h_rank$cum_non_resp / sum(h_rank$cnt_non_resp),2);
h_rank$ks <- abs(h_rank$cum_rel_resp - h_rank$cum_rel_non_resp);
library(scales)
h_rank$rrate <- percent(h_rank$rrate)
h_rank$cum_rel_resp <- percent(h_rank$cum_rel_resp)
h_rank$cum_rel_non_resp <- percent(h_rank$cum_rel_non_resp)
h_rank## deciles cnt cnt_resp cnt_non_resp rrate cum_resp cum_non_resp
## 1: 10 600 355 245 59% 355 245
## 2: 9 656 206 450 31% 561 695
## 3: 8 617 84 533 14% 645 1228
## 4: 7 554 58 496 10% 703 1724
## 5: 6 675 28 647 4% 731 2371
## 6: 5 1055 41 1014 4% 772 3385
## 7: 4 855 9 846 1% 781 4231
## 8: 2 988 14 974 1% 795 5205
## cum_rel_resp cum_rel_non_resp ks
## 1: 45% 5% 0.40
## 2: 71% 13% 0.58
## 3: 81% 24% 0.57
## 4: 88% 33% 0.55
## 5: 92% 46% 0.46
## 6: 97% 65% 0.32
## 7: 98% 81% 0.17
## 8: 100% 100% 0.00ROC Curve
library(ROCR)
pred <- prediction(RFDF.holdout$predict.score[,2],RFDF.holdout$TARGET)
perf <- performance(pred, "tpr", "fpr")
plot(perf)
KS value
KS <- max(attr(perf, 'y.values')[[1]]-attr(perf, 'x.values')[[1]])
KS## [1] 0.5860439Gini coefficient
library(ineq)
gini = ineq(RFDF.holdout$predict.score[,2], type="Gini")
gini## [1] 0.7193523